Chapter 26 - Machine learning in bioinformatics

Abstract

This chapter highlights the power of machine learning approaches in bioinformatics and biological sciences. It is an adaptive mechanism that allows computers to improve from experiences and examples. It is therefore a methodological discipline that offers processing capabilities with intelligent information for managing real-life information of one or another. It usually constructs a mathematical model of sample datasets, known as training datasets, to make predictions or decisions on the target datasets. As in the era of omics, a huge amount of biological data being generated every day that requires machine learning approaches for useful decisions and predictions. This will be useful for reducing experimental cost and time. Machine learning approaches play a crucial role in a different area of bioinformatics, including gene findings and genome annotation, protein structure prediction, gene expression analysis, complex interaction modeling in biological systems, drug discovery, text mining, and digital image processing. The chapter also presents recent advances and limitations of machine learning algorithms used for bioinformatics.